HCUP NIS - Healthcare Cost and Utilization Project, National (Nationwide) Inpatient Sample

General description

1. Database primary purpose - Guide researchers and policymakers on making national estimates of health care utilization, cost, quality, and outcomes. The NIS contains information on all hospital stays, regardless of expected payer for the hospital stay.
2. Overall data type - Health outcomes
3. Dataset type - Cross-sectional
4. Data source - Claims
5. Data level - Patient level, but includes a Hospital-level file
6. Geographic location of the data collection sites - United States
7. Sponsor, manager, or home institution - Agency for Healthcare Research and Quality's (AHRQ)
8. Date range - 2017 - 2019
9. Geolocation data - hospital region, hospital, and patient urban-rural code
10. Dates - Year; admission weekday and month; discharge quarter
11. Hospital identifiers - NIS hospital number
12. Financial variables - Contains charge information and provides supplemental files containing cost-to-charge ratios
13. Clinical areas of interest - all
14. Variables that are uniquely present in this dataset - NIS is the largest publicly available all-payer inpatient healthcare database designed to produce U.S. regional and national estimates of inpatient utilization, access, cost, quality, and outcomes. Its adjusted estimates cover more than 97 percent of the U.S. population.
15. Database caveats and limitations - Not all data elements are available for every State, and not for every year. Also, not all data elements are uniformly coded across states.
16. Other - The NIS is sampled from the State Inpatient Databases (SID), including all inpatient data that are currently contributed to HCUP.

Applicable methods

1. Association methods, such as multivariable logistic regression (1, 2), hierarchical regression (3, 4), linear regression models (5, 6)
2. Clustering (7)
3. Exploratory analysis (8, 9, 10)
4. Inferential test (11)
5. Interrupted time series (12)
6. Machine learning techniques (13, 14, 15)
7. Propensity score (, 16, 17)
8. Sensitivity analysis (18)
9. Time series (19)

High-impact designs

• Compare NIS and State Inpatient Databases (SID) data to conduct state health policy research (20)

• Evaluate patient demographics and clinical characteristics of specific diagnosis-related hospitalizations (21, 22)

• Compare childbirth costs for adolescents and adults (23)

• Describe published summary data on selected safety measures (24)

• Assess whether/how outcomes changed after new policy implementation (12, 25)

• Racial disparities and surgical outcomes (26)

• Examine if elective surgery on diagnosis would prevent the development of complications and avoid the risk of emergency surgery (27)

• Assess the economic impact of obesity on hospital costs associated with surgical procedures (28)

Data dictionary

To access the data dictionary, click here.

Variable categories

1. Patient demographics (e.g., age, sex, race, ethnicity, median household income)
2. Hospital discharge records (e.g., diagnosis-related group (DRG) in use on discharge date, number of diagnoses on discharge, LOS)
3. Charges (e.g., expected payer, total charges)
4. Diagnosis codes (including injury type)
5. Procedure codes
6. Hospital characteristics (e.g., control/ownership, bed size, STRATA, number of hospital discharges, and hospital markets file defined by geopolitical boundaries, fixed radius, variable radius, and patient flow)
7. Disposition at discharge (e.g., discharge to home, transferred to [type of facility], left against medical advice)
8. Died during hospitalization
9. Severity and comorbidity measures